Immigrants in the United States: End Notes

Back to Table of Contents


1 It does not include those born abroad of American parents or those born in outlying territories of the United States such as Puerto Rico.

2 For the post-1980 immigrant population, the Department of Homeland Security estimates a 5 percent undercount in the ACS. See the DHS publication "Estimates of the Unauthorized Immigrant Population Residing in the United States: January 2011", at http://www.dhs.gov/xlibrary/assets/statistics/publications/ois_ill_pe_20.... The Pew Hispanic Center comes to a similar conclusion in their analysis of the Current Population Survey. See "Trends in Unauthorized Immigration: Undocumented Inflow Now Trails Legal Inflow", October 2008, at http://pewhispanic.org/files/reports/94.pdf.

3 See, for example, "Immigration and Economic Stagnation: An Examination of Trends 2000 to 2010", November 2010, http://cis.org/highest-decade. See also "Homeward Bound: Recent Immigration Enforcement and the Decline in the Illegal Alien Population", July 2008, http://cis.org/trends_and_enforcement; and "A Shifting Tide: Recent Trends in the Illegal Immigrant Population", July 2009, http://www.cis.org/IllegalImmigration-ShiftingTide.

4 The Pew Hispanic Center assumes a 5.2 percent undercount of the total foreign-born in the 2005 CPS. See Figure 3, p. 4 in their March 2006 estimate of the illegal population, at http://pewhispanic.org/files/reports/61.pdf. Pew bases its 5.2 percent estimate on work done by Passel, Van Hook, and Bean. Their paper, "Narrative Profile with Adjoining Tables of Unauthorized Migrants and Other Immigrants, Based on Census 2000: Characteristics and Methods", was done for Sabre Systems as part of a contract with the Census Bureau.

5 See End Note 2 for a discussion of the ACS and the decennial census.

6 Starting in 1970, the Census began to ask about year of arrival, so arrival by decade is available for the 1960s onward. Administrative data on legal immigration goes back to 1820 and shows that no decade comes close to the nearly 14 million immigrants who arrived from 2000 to 2010.

7 If we line up the arrival data with job growth and compare January 1990 to April of 2000 (the date of the Census), job growth was 22 million. If we compare January 2000 to July 2010 (the control data for the ACS) we find a decline of 425,000 jobs. In a previous study we reported a net decline of over a million jobs from 2000 to 2010 and a net gain of some 21 million from 1990 to 2000. That study, "Immigration and Economic Stagnation: An Examination of Trends 2000 to 2010", http://www.cis.org/highest-decade, used the year of arrival data from the Current Population Survey, which is taken in March. The Census is taken in April and the ACS is weighted to reflect the population in July. Although the months compared in that earlier study were different, resulting in somewhat different job figures, the basic conclusion is exactly the same. Immigration was higher 2000 to 2010 compared to 1990 to 2000, even though the economy was fundamentally different in each decade. Historical data from 1994 to the present can be found at http://www.bls.gov/schedule/archives/empsit_nr.htm#2010. Figures for January 1990 can be found in the November 1990 issue of Monthly Labor Review, at http://www.bls.gov/opub/mlr/1990/11/rpt1full.pdf.

8 For example, new arrivals were very high in 2000 based on the year of arrival data (Figure 3) from the 2001 ACS, but there was little growth in the foreign-born between 2000 and 2001 (Figure 2). However, the 2000 total foreign-born number is from the decennial census, while the 2001 total and the arrival data for that year are from the ACS. The ACS was not fully implemented in 2001, and that survey differs from the Census in other ways that may explain why high levels of new immigration in 2000 did not produce high growth in the immigrant population. Another seeming incongruity is the high growth from 2003 to 2004 of 1.4 million (Figure 2), yet Figure 3 shows new arrivals were only 1.2 million in 2003. Also there was very high growth (1.9 million) from 2004 to 2005 even though new arrivals in 2004 were 1.35 million. However, it must be remembered that the ACS reflects a July 1 estimate of the U.S. population, including the foreign-born. So individual year of arrival data, which correspond to calendar year, do not directly compare to growth July 1 to July 1 of each year. Moreover, the ACS was not fully implemented until 2005 and individuals in group quarters were not included in the ACS until 2006. These factors also impact year-to-year comparisons. All of these issues create important breaks in the continuity of the data.

9 Because the public-use ACS files report individual year of arrival, it is easy to calculate the average length of time immigrants have been in the country in 2010.

10 The figures for Tables 3 and 4 come from a Center for Immigration Studies analysis of the 2010 public-use file of the American Community Survey, which shows 39,916,875 immigrants. However the public-use ACS files, while designed to be representative of the ACS data used internally by the Census Bureau, do not exactly match the internal file, hence the 39,000 (0.01 percent) difference in the totals found in Tables 3 and 4 compared with the totals in Tables 1, 2, and 5, which come from the internal data used by the Census Bureau in American FactFinder.

11 Countries that can be identified in the public-use ACS file and for which there were actually respondents by region are as follows: Central America: Belize, Costa Rica, El Salvador, Guatemala, Honduras, Nicaragua, and Panama. South America: Argentina, Bolivia, Brazil, Chile, Colombia, Ecuador, Guyana, Paraguay, Peru, Uruguay, Venezuela, and South America not specified. South Asia: Afghanistan, India, Bangladesh, Pakistan, Sri Lanka, and Nepal. East Asia: China, Hong Kong, Taiwan, Japan, Korea, Cambodia, Indonesia, Laos, Malaysia, Philippines, Singapore, Thailand, Vietnam, Burma, East Asia and Asia not specified. Europe: Denmark, Finland, Iceland, Norway, Sweden, United Kingdom, Ireland, Belgium, France, Netherlands, Switzerland, Albania, Greece, Macedonia, Italy, Portugal, Spain, Austria, Bulgaria, Czechoslovakia, Slovakia, Czech Republic, Germany, Hungary, Poland, Romania, Yugoslavia, Croatia, Bosnia, Estonia, Latvia, Lithuania, Byelorussia, Moldova, Ukraine, Armenia, Georgia, Russia, USSR not specified, and Europe not specified. Caribbean: Bermuda, Cuba, Dominican Republic, Haiti, Jamaica, Antigua-Barbuda, Bahamas, Barbados, Dominica, Grenada, St. Kitts-Nevis, St. Lucia, St. Vincent, Trinidad and Tobago, and Caribbean and West Indies not specified. Middle East: Azerbaijan, Kazakhstan, Uzbekistan, Iran, Iraq, Israel/Palestine, Jordan, Kuwait, Lebanon, Saudi Arabia, Syria, Turkey, Yemen, Algeria, Egypt, Morocco, Sudan, and North Africa not specified. Sub-Saharan Africa: Cape Verde, Ghana, Guinea, Liberia, Nigeria, Senegal, Sierra Leone, Ethiopia, Kenya, Somalia, Tanzania, Uganda, Zimbabwe, Eritrea, Cameroon, South Africa, and Africa and Western and Eastern Africa not specified.

12 This is based on the number of immigrants in the 2010 ACS who indicated that they arrived in 2000.

13 Based on the child's age, the March 2010 Current Population Survey shows that there were 8.981 million children born in the United States to immigrant fathers during the prior decade. Of these births 2.257 million were to fathers who arrived in 2000 or later. All of these children were living in the United States in 2010.

14 It may also be helpful to think about the limitations of using just growth in the immigrant population by considering the fact that if, say, one million immigrants enter the country each year, at some point one million immigrants will eventual die a year, assuming no out-migration. This would mean that the arrival of one million new immigrants roughly equaled deaths and thus there is no growth in the foreign-born population. But of course the U.S. population would in fact be much larger with the arrival of one million new immigrants regardless of mortality.

15 For a discussion of the decline in immigrant education relative to natives, see "The Slowing Progress of Immigrants: An Examination of Income, Home Ownership, and Citizenship, 1970-2000", at www.cis.org/sites/cis.org/files/articles/2001/back401.html.

16 This figure refers to persons 18 or older who are in the workforce. To be in the workforce one has to be either employed or actively looking for work. Persons actively looking for work are considered unemployed.

17 The median figures in Table 7 and all subsequent tables, including those for households, are calculated using the Census Bureau method of grouping data into $2,500 cells. While the median figures in this Backgrounder very closely match median figures published by the Census Bureau, they may not exactly match in all cases because the bureau top-codes income figures in the public-use files of the CPS.

18 The report, "Immigration in an Aging Society Workers, Birth Rates, and Social Security", can be found at www.cis.org/sites/cis.org/files/articles/2005/back505transcript.html.

19 See p. 21 of the Census Bureau's "Methodology and Assumptions for the Population Projections of the United States: 1999 to 2100", at www.census.gov/population/www/documentation/twps0038.pdf.

20 The report, "100 Million More: Projecting the Impact of Immigration on the U.S. Population, 2007 to 2060", can be found at www.cis.org/sites/cis.org/files/articles/2007/back707.html.

21 It should be noted that the unemployment rate cannot be calculated by comparing the difference between those with a job and those in the labor force because the Bureau of Labor Statistics calculates unemployment by dividing those actively looking for a job by the labor force. In contrast, the share holding a job or the share in the labor force are based on the entire 18- 65-year-old population.

22 Like official U.S. government poverty statistics, the poverty statistics in this report do not include persons under age 15 unrelated to the household head. This excludes about 400,000 children, who are mostly in foster care.

23 Figures for natives exclude the young (under 18) U.S.-born children of immigrant fathers.

24 Countries that can be identified in the public-use CPS files and for which there were actually respondents by region are as follows: Europe: Albania, Austria, Belgium, Bulgaria, Czechoslovakia, Denmark, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Netherlands, Norway, Poland, Portugal, Romania, Spain, Sweden, Switzerland, United Kingdom, Yugoslavia, Czech Republic, Slovakia, Bosnia & Herzegovina, Croatia, Macedonia, Serbia, Latvia, Lithuania, Armenia, Belarus, Georgia, Moldova, Ukraine, Russia/USSR not specified, Kosovo, Cyprus, and Europe not specified. South Asia: Afghanistan, Bangladesh, India, Nepal, Pakistan, and Sri Lanka. East Asia: Myanmar, Cambodia, China, Indonesia, Japan, Korea, Malaysia, Philippines, Singapore, Thailand, Vietnam, and Asia not specified. Middle East: Azerbaijan, Iran, Iraq, Israel, Jordan, Kuwait, Laos, Lebanon, Saudi Arabia, Syria, Turkey, Uzbekistan, Yemen, Algeria, Egypt, Morocco, and Sudan. Central America: Belize, Costa Rica, El Salvador, Guatemala, Honduras, Nicaragua, and Panama. Caribbean: Bermuda, Antigua and Barbuda, Bahamas, Barbados, Cuba, Dominica, Dominican Republic, Grenada, Haiti, Jamaica, St. Kitts-Nevis, St. Lucia, St. Vincent and the Grenadines, Trinidad and Tobago, and West Indies not specified. South America: Argentina, Bolivia, Brazil, Chile, Colombia, Ecuador, Guyana, Paraguay, Peru, Uruguay, Venezuela, and South America not specified. Sub-Saharan Africa: Cameroon, Cape Verde, Ethiopia, Eritrea, Ghana, Kenya, Liberia, Nigeria, Senegal, Sierra Leone, Somalia, South Africa, Tanzania, Uganda, Zimbabwe, and Africa not specified.

25 Sarita A. Mohanty, Steffie Woolhandler, David U. Himmelstein, Susmita Pati, Olveen Carrasquillo, and David H. Bor, "Health Care Expenditures of Immigrants in the United States: A Nationally Representative Analysis", American Journal of Public Health, Vol. 95, No. 8.

26 Using the revised weights from the March 2000 CPS shows 40.23 million uninsured in 1999, for a growth of 9.68 million compared to the March 2011 CPS. The 2011 CPS shows 6.2 million uninsured immigrants in the country who arrived in 2000 or later and that 385,000 of those same immigrants have U.S.-born children who are uninsured.

27 Figures for immigrants include the U.S.-born children (under 18) of immigrant fathers. Figures for natives exclude these children.

28 See, for example, Figures 20-1, 20-2, and 20-3 in "Profiles of the Foreign-born Population in the United States 2000", U.S. Government Printing Office. Dianne A. Schmidley, Series P23-206.

29 See Appendix F in Measuring the Effect of Benefits and Taxes on Income and Poverty: 1992, U.S. Census Bureau, for a discussion of under-reporting of income and receipt of redistribution programs.

30 The Census Bureau released the 2011 CPS without figures for the EITC and ACTC, thus the figures for the Credit are from the 2010 CPS. The Additional Child Tax Credit can also be referred to as the refundable portion of the Child Tax Credit. Table 12 reports those who are eligible for cash from the government, not just a refund of money they paid as taxes.

31 Use of welfare for both immigrants and natives in the CPS is understated because people forget about use when answering the survey. This is particularly true of Medicaid and cash assistance programs, which administrative data show are somewhat higher than the numbers found in the CPS.

32 The primary refugee-sending countries that can be identified in the CPS are Albania, the former Yugoslavia, the former USSR, Afghanistan, Iraq, Cuba, Ethiopia, Eritrea, Liberia, Sierra Leone, Somalia, Sudan, Vietnam, Cambodia, Laos, and Nicaragua.

33 See "Measuring Overcrowding in Housing, 2007", from the U.S. Department of Housing and Urban Development Office of Policy Development and Research, at www.huduser.org/publications/pdf/Measuring_Overcrowding_in_Hsg.pdf. It is worth noting that for reasons that are not entirely clear there has been a significant decline in household overcrowding. There is debate about how much of this decline is due to changes in data collection and how much is a real decline. But this issue does not affect the analysis in this report because we are only examining figures for a single year.

34 To calculate household size we exclude all those in group quarters such as prisons, nursing homes, and college dorms.

35 Calculations of home ownership exclude those in group quarters.

36 All immigrants in the ACS are asked what year they came to the United States. For the purposes of this analysis, 20 years is defined in the 2010 ACS as having entered the country in 1989, 1990, or 1991. We average three years together in order to obtain a more robust estimate.

37 The March 2010 Current Population Survey (CPS) shows that 20.5 percent of school-age children have an immigrant father and the March 2011 CPS for 2010 shows 21.3 percent of the school-age population have an immigrant father. Both the ACS and CPS produce very similar results, but we use the ACS because, unlike the March CPS, it distinguishes between public and private schools. Another advantage of the ACS is that it includes a question on language, which is an important issue in public education.

38 As is the case in other tables, the figures for natives exclude the U.S.-born children under age 18 of immigrant fathers.

39 Poverty, earnings, health insurance, and welfare use are based on the March 2011 CPS. Those in the country five or fewer years arrived in 2006 or later. Those in the United States 20 years arrived from 1988 to 1993. By 2011, on average, these immigrants had lived in the country for 20.5 years. Coding the data in this way by year of arrival is necessary given the way respondents are grouped by the Census Bureau in the public-use files. Also, grouping 1988 to 1993 provides for a larger sample. Homes ownership is based on the ACS, and those in the country for 20 years arrived in 1989, 1990, or 1991.

40 It is not possible to identify generations beyond the third with the CPS, so all those with two U.S.-born parents constitute the "third generation-plus", regardless of where their grandparents were born.

41 The 2010 ACS shows that 55 percent of immigrant women who had a child in the prior year were Hispanics. The 2011 CPS shows that, of U.S.-born children in the United States with an immigrant father, 59 percent are Hispanic.

42 Brian Duncan and Stephen J. Trejo, "Ancestry vs. Ethnicity: The Complexity and Selectivity of Mexican Identification in the United States", pp. 31-66 in Amelie F. Constant, Konstantinos Tatsiramos, and Klaus F. Zimmermann, eds., Ethnicity and Labor Market Outcomes (Research in Labor Economics, Volume 29), 2011. And "Who Remains Mexican? Selective Ethnic Attrition and the Intergenerational Progress of Mexican Americans", pp. 285-320 in David Leal and Stephen Trejo, eds., Latinos and the Economy: Integration and Impact in Schools, Labor Markets, and Beyond, New York: Springer.

43 A modest share of the "third generation-plus" are decedents of people living in Texas or the American southwest when it was part of the Spanish Empire or Mexico.

44 Of second-generation adult Hispanics, 9.5 percent used Medicaid compared to 14 percent of third-generation Hispanics.

45 The median earnings of foreign-born Hispanics are $20,727, or 62 percent that of non-Hispanic natives ($33,435); for second-generation Hispanics it is $24,390, or 73 percent that of non-Hispanic natives; and it is $26,926, or 81 percent that of non-Hispanic natives for third-generation Hispanics.

46 The average age of adult second-generation Hispanics in the 2011 CPS is 33.8 and for the third generation-plus, it is 41 years. This compares to the average age of 47.2 years for adult U.S.-born non-Hispanics.

47 See for example, Patricia Gandara and Frances Contreras, The Latino Education Crisis: The Consequences of Failed Social Policies, Harvard University Press, 2009; and Carola Suárez-Orozco, Marcelo M. Suárez-Orozco and Irina Todorova, Learning a New Land: Immigrant Students in American Society, 2008.

48 In his work, Harvard economist George Borjas has emphasized that large initial differences in human capital among the immigrant generation can persist through into following generations. See for example, George J. Borjas, "The Intergenerational Mobility of Immigrants", Journal of Labor Economics,1993; and George Borjas, "Working Paper 12088", in Making It in America: Social Mobility in the Immigrant Population, National Bureau of Economic Research, 2006, http://www.nber.org/papers/w12088. See also Stephen J. Trejo, "Why Do Mexican Americans Earn Low Wages?" Journal of Political Economy, Vol. 105, No. 6, December 1997; and Gretchen Livingston and Joan R. Kahn, "An American Dream Unfulfilled: The Limited Mobility of Mexican Americans", Social Science Quarterly, Vol. 83, No. 4, December 2002.

49 As will be recalled from the discussion on page 29, 50.2 percent are on Medicaid or uninsured based on the 2011 CPS alone. This is very similar to the 49 percent shown when using a combined sample of the 2010 and 2011 CPS.

50 To distinguish legal from illegal immigrants in the survey, this report uses citizenship status, year of arrival in the United States, age, country of birth, educational attainment, sex, receipt of welfare programs, receipt of Social Security, veteran status, and marital status. We use these variables to assign probabilities to each respondent. Those individuals who have a cumulative probability of 1 or higher are assumed to be illegal aliens. The probabilities are assigned so that both the total number of illegal aliens and the characteristics of the illegal population closely match other research in the field, particularly the estimates developed by the Department of Homeland Security/legacy INS, the Urban Institute, and the Pew Hispanic Center. This method is based on some very well-established facts about the characteristics of the illegal population. For example, it is well known that illegal aliens are disproportionately young, male, unmarried, under age 40, have few years of schooling, etc. Thus, we assign probabilities to these and other factors in order to select the likely illegal population. In some cases, we assume that there is no probability that an individual is an illegal alien.

51 Both DHS estimates and Pew Hispanic estimates assume this level of undercount in the Census Bureau data.

52 The Department of Homeland Security (DHS) estimate of 11.5 million illegal immigrants in January 2011 can be found at http://www.dhs.gov/xlibrary/assets/statistics/publications/ois_ill_pe_20.... That estimate includes an adjustment for those missed in Census Bureau data. The Pew Hispanic Center has estimated a 11.2 million illegal immigrant population as of March 2010 based on the CPS. This includes an adjustment for those missed by the survey. The Pew report can be found at http://www.pewhispanic.org/2011/02/01/unauthorized-immigrant-population-.... Older studies by the INS and Census Bureau are also available. The INS report that found seven million illegal aliens in 2000 and an annual increase of about 500,000 can be found at http://www.dhs.gov/xlibrary/assets/statistics/publications/Ill_Report_12.... The Census Bureau estimate of eight million illegals in 2000 can be found at www.census.gov/dmd/www/ReportRec2.htm. (Appendix A of Report 1 contains the estimates.)

53 The 2011 CPS by itself, reported in Table 10, showed 13.5 percent in poverty.

54 To identify illegal immigrants in the CPS we assume that a household head who reports living in public or subsidized housing cannot be an illegal immigrant. This does not mean that illegal immigrants do not live in subsidized housing. It simply means they cannot be the name on the lease, which is one of the things that defines the household head.

55 To calculate median income in the way that the Census Bureau does, it is necessary to group data into cells. But the number of illegal alien households, which is much smaller than the number of illegal alien individuals, is not large enough in most states to do this and still produce reliable results. In contrast, a mean or average figure does not require the grouping of data so it is possible to calculate average income for illegal immigrant households, even in smaller states.

56 As already discussed, if we have overestimated welfare use for households headed by illegal aliens, then legal immigrants, particularly the unskilled, must have even higher welfare use rates than reported here. This would mean that legalization would be even more costly because the difference between what illegals currently use and what they would use once legalized is even larger. This has to be the case mathematically because immigrant households accessing the welfare system can only be legal immigrants or illegal aliens and we simply take the welfare use rates for the foreign-born as reported in the CPS. We do not impute welfare use or change the share using welfare for the foreign-born in any way.

57 It should be pointed out that adult, less-educated legal immigrants in the CPS have lived in the United States longer than the average illegal immigrant. Over time income rises. The estimates for less-educated legal immigrants reflect this fact. Thus, less-educated legal immigrants in the CPS have higher incomes than would be expected for legalized illegal immigrants after any amnesty, at least initially.